In FM systems, it is well known that the noise at the output of a discriminator is inversely proportional to the signal. That is, with no signal, or very low signal, the discriminator output is substantially all noise. As the strength of the signal increases the noise output from the discriminator is reduced. This phenomenon is utilized in many quieting or squelch systems as well as in many voting systems (a system incorporating a plurality of receivers and wherein the receiver providing the strongest signal is utilized).
In the prior art a great variety of signal strength or quality detectors are provided, most of which provide a direct indication of the amplitude of the noise at the output of the discriminator or in some way compare the noise to the audio at the output of the discriminator. The bandwidth of the discriminator is generally equal to the width of the communications channel (i.e. 10 kHz) while the audio spectrum is generally substantially smaller (i.e. 300 to 3000 kHz). Thus, there is substantial bandwidth above the audio spectrum to measure the amplitude of the noise and determine the strength of the signal being received, if any.
In systems wherein the audio being transmitted is voice signals the noise above the audio spectrum is a relatively reliable indication of signal strength. However, in tone signalling systems harmonics of the tones are prevalent throughout the bandwidth of the discriminator and the harmonic of a particular tone can appear at the exact point at which the noise is being measured so that the measurement will not be a direct indication of signal strength or quality.